[ECOS] powerpc hal structure

[ECOS] powerpc hal structure

[Thomas Koeller]

Hi,

I am just about to begin a new ecos port for a powerpc target, and when
I took a look at what already exists, some questions arose.

I noticed that all the existing platform hal packages use CYGBLD_HAL_TARGET_H
to define the variant header file path, shouldn't they be using
CYGBLD_HAL_VARIANT_H for this purpose? OTOH, the arch hal seems
to expect this use of CYGBLD_HAL_TARGET_H, as it uses it in powerpc.ld.

Since my CPU (Motorola MPC8240) is a variant of the 603, I originally assumed
that I could build upon the mpc60x hal. But looking at that I got the impression
that it is mostly a dummy, containing just enough functionality to run the
simulator, and besides that, building a variant hal on to of another one wouldn't
fit the arch/var/platform hal trinity. So I guess it's probably more appropriate
to support the mpc824x by a variant hal of its own, cloning the existing mpc60x
hal instead of extending it. Or should I add my code to that package, probably
encapsulated as a cdl_component? This would require replacing dummy functions
with real ones, which would probably break the simulator or require a lot of
conditonal code, to retain the dummies for the simulator's sake.

tk

-- 

Thomas Koeller, Software Development

Basler Vision Technologies
An der Strusbek 60-62
22926 Ahrensburg
Germany

Phone	+49 (4102) 463-390
Fax	+49 (4102) 463-46390

mailto:thomas.koeller@baslerweb.com
http://www.baslerweb.com 



-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss